Audio samples for CHiVE

Audio samples for CHiVE: Varying Prosody in Speech Synthesis with a Linguistically Driven Dynamic Hierarchical Conditional Variational Network, V. Wan, C.-a. Chan, T. Kenter, J. Vit, and R. A. Clark, Proceedings of the Thirty-sixth International Conference on Machine Learning (ICML 2019), 2019.

@inproceedings{vwan2019chive,
  title = {CHiVE: Varying Prosody in Speech Synthesis with a Linguistically Driven Dynamic Hierarchical Conditional Variational Network},
  author={Vincent Wan and Chun-an Chan and Tom Kenter and Jakub Vit and Rob Clark},
  booktitle={Thirty-sixth International Conference on Machine Learning (ICML 2019)},
  pages = {3331--3340}
  year = {2019}
}

Samples that go with the main experiment

These samples go with the side-by-side experiments (i.e., Table 1 in the paper).

Samples of further evaluations

These samples go with the MOS test (i.e., Section 4.2, Table 2 in the paper).
These samples are of randomly sampled pitch contours (i.e., Section 4.2, Figure 4 in the paper).

Samples of prosody transfer

These samples go with the prosody transfer illustrations (i.e., Section 4.3, Figures 5 and 6 in the paper, and some additional ones).